Omid-Nejati / BEFUnet

A Hybrid CNN-Transformer Architecture for Precise Medical Image Segmentation

BEFUnet: A Hybrid CNN-Transformer Architecture for Precise Medical Image Segmentation

📕 [arxiv]

🎉 🎉 🎉 News

• 2024/03/20 First release.

Train & Test --- Synapse Dataset

Please go to "Colab_BEFUnet.ipynb" for complete detail on dataset preparation and Train/Test procedure or follow the instructions below.

Usage

This code has been implemented in python language using Pytorch library and tested in ubuntu OS, though should be compatible with related environment. following Environement and Library needed to run the code: Python 3, Pytorch

Installation

1. Run the following code to install the Requirements.

   pip install -r requirements.txt

2. Run the below code to train BEFUnet on the synapse dataset.

   python train.py --root_path ./data/Synapse/train_npz --test_path ./data/Synapse/test_vol_h5  --model_name BEFUnet --batch_size 10 --eval_interval 20 --max_epochs 500 

3. Run the below code to test BEFUnet on the synapse dataset.

   python test.py --test_path ./data/Synapse/test_vol_h5 --model_name BEFUnet --is_savenii --model_weight 

Query

All implementations are done by Omid Nejati Manzari. For any query please contact us for more information.

[e-mail]

Acknowledgement

We borrowed the code from Swin Transformer and PiDinet. Thanks for their wonderful works.

Citation

If you find this project helpful for your research, please consider citing the following BibTeX entry.

@article{manzari2024befunet,
  title={BEFUnet: A Hybrid CNN-Transformer Architecture for Precise Medical Image Segmentation},
  author={Manzari, Omid Nejati and Kaleybar, Javad Mirzapour and Saadat, Hooman and Maleki, Shahin},
  journal={arXiv preprint arXiv:2402.08793},
  year={2024}
}